DISENTANGLING THE MECHANISMS UNDERLYING URBAN FORESTS AS NATURE-BASED SOLUTIONS ON URBAN HEAT ISLANDS

My PhD research delves into the Urban Heat Island (UHI) phenomenon across various scales and contexts, focusing on the interplay between urban features and land surface temperature (LST) to inform sustainable urban planning and climate resilience strategies. The research unfolds over three interconnected studies, each contributing to a comprehensive understanding of UHI’s global expansion, its key drivers, and potential mitigation approaches. The first study, The Unrelenting Global Expansion of the Urban Heat Island Over the Last Century, provides a historical perspective on UHI’s spread from 1901 to 2022, analyzing over 6,000 publications and tracking UHI trends across 1,726 cities globally. This study reveals a consistent increase in the number of cities affected by UHI, largely influenced by rising urbanization and global temperature. Notably, it highlights the frequency of new cities experiencing UHI effects for the first time, underscoring the global relevance of UHI and the urgent need for large-scale, climate-resilient urban planning strategies to mitigate its adverse effects. The second study, Exploring the Non-Linear Impacts of Urban Features on Land Surface Temperature Using Explainable Artificial Intelligence, focuses on Beijing, China, using advanced machine learning models combined with Explainable Artificial Intelligence (XAI). This approach helps quantify the influence of urban features—such as elevation, impervious surface compactness, and vegetation density—on LST. By uncovering complex, non-linear interactions, this study reveals that LST drivers vary substantially within different parts of the city. For example, the cooling effect of impervious surfaces diminishes beyond a certain density, while tree height’s influence on LST is more effective in specific city zones. These insights inform tailored, data driven interventions in urban planning to address UHI in dense metropolitan settings. The third study, Understanding the Coupling Effect of Multiple Urban Features on Land Surface Temperature in Europe, shifts focus to 780 cities in eight European macro regions, evaluating how both ecological and structural features interact to impact LST. Using Random Forest models and Generalized Additive Models (GAMs), this study finds that ecological attributes like tree height and evapotranspiration have a significant cooling effect in warmer climates such as the Mediterranean and Turkey. In cooler Relazione triennio regions like Scandinavia, however, the impact of these ecological features is less pronounced, and other factors, such as coastal proximity or urban structure, take precedence. These findings stress the importance of region-specific urban planning strategies that integrate both natural and built environments to effectively moderate urban temperatures. Together, these studies present a multi-scale analysis of UHI, from global patterns to detailed regional and local evaluations. My PhD research emphasizes that managing urban heat requires an integrated approach that considers the unique environmental characteristics of each city or region. By examining non-linear interactions and identifying dominant cooling and heating factors, the studies reveal that no single approach suits all cities; instead, a data-driven, context-sensitive approach is essential for effective UHI mitigation. These findings have wide-reaching implications for urban planning and policy-making. The research demonstrates that climate resilience in urban areas can be strengthened by strategic interventions focused on ecological attributes, such as preserving and enhancing natural vegetation, and on structural urban elements, such as managing impervious surface density and regulating building density and height.